sh: Add \u/\U support (in $'...') for UTF-8.

Because we have no iconv in base, support for other charsets is not
possible.

Note that \u/\U are processed using the locale that was active when the
shell started. This is necessary to avoid behaviour that depends on the
parse/execute split (for example when placing braces around an entire
script). Therefore, UTF-8 encoding is implemented manually.

1 files changed, 23 insertions, 0 deletions

diff --git a/bin/sh/parser.c b/bin/sh/parser.c
index 5133d67..cc1860d 100644
--- a/bin/sh/parser.c
+++ b/bin/sh/parser.c
@@ -1219,6 +1219,29 @@ readcstyleesc(char *out)
 		  if (v == 0 || (v >= 0xd800 && v <= 0xdfff))
 			  synerror("Bad escape sequence");
 		  /* We really need iconv here. */
+		  if (initial_localeisutf8 && v > 127) {
+			  CHECKSTRSPACE(4, out);
+			  /*
+			   * We cannot use wctomb() as the locale may have
+			   * changed.
+			   */
+			  if (v <= 0x7ff) {
+				  USTPUTC(0xc0 | v >> 6, out);
+				  USTPUTC(0x80 | (v & 0x3f), out);
+				  return out;
+			  } else if (v <= 0xffff) {
+				  USTPUTC(0xe0 | v >> 12, out);
+				  USTPUTC(0x80 | ((v >> 6) & 0x3f), out);
+				  USTPUTC(0x80 | (v & 0x3f), out);
+				  return out;
+			  } else if (v <= 0x10ffff) {
+				  USTPUTC(0xf0 | v >> 18, out);
+				  USTPUTC(0x80 | ((v >> 12) & 0x3f), out);
+				  USTPUTC(0x80 | ((v >> 6) & 0x3f), out);
+				  USTPUTC(0x80 | (v & 0x3f), out);
+				  return out;
+			  }
+		  }
 		  if (v > 127)
 			  v = '?';
 		  break;